Implementing a Task
Execute the implementation plan, track progress, and complete the task.
Announce: "Using kn-implement for task [ID]."
Core principle: CHECK AC ONLY AFTER WORK IS DONE.
Step 1: Review Task
json
mcp__knowns__get_task({ "taskId": "$ARGUMENTS" })
If task status is "done" (reopening):
json
mcp__knowns__update_task({
"taskId": "$ARGUMENTS",
"status": "in-progress",
"appendNotes": "Reopened: <reason>"
})
mcp__knowns__start_time({ "taskId": "$ARGUMENTS" })
Verify: plan exists, timer running, which ACs pending.
Step 2: Check Templates
json
mcp__knowns__list_templates({})
If template exists → use it to generate boilerplate.
Step 3: Work Through Plan
For each step:
- •Do the work
- •Check AC (only after done!)
- •Append note
json
mcp__knowns__update_task({
"taskId": "$ARGUMENTS",
"checkAc": [1],
"appendNotes": "Done: brief description"
})
Step 4: Handle Scope Changes
Small: Add AC + note
json
mcp__knowns__update_task({
"taskId": "$ARGUMENTS",
"addAc": ["New requirement"],
"appendNotes": "Scope: added per user"
})
Large: Stop and ask user.
Step 5: Validate & Complete
- •Run tests/lint/build
- •Validate to catch broken refs:
json
mcp__knowns__validate({})
- •Add implementation notes (use
appendNotes, NOTnotes!) - •Stop timer + mark done
json
mcp__knowns__stop_time({ "taskId": "$ARGUMENTS" })
mcp__knowns__update_task({
"taskId": "$ARGUMENTS",
"status": "done"
})
Step 6: Extract Knowledge (optional)
If patterns discovered: /kn-extract $ARGUMENTS
Checklist
- • All ACs checked
- • Tests pass
- • Validated (no broken refs)
- • Notes added
- • Timer stopped
- • Status = done
Red Flags
- •Checking AC before work done
- •Skipping tests
- •Skipping validation
- •Using
notesinstead ofappendNotes - •Marking done without verification